Zebra Medical Vision (Zebra-Med), an Israel-based deep-learning medical imaging analytics company, said that its software will be deployed by India’s Apollo Hospitals Group to help its doctors in efficient and accurate diagnoses of COVID-19, by leveraging the former’s0 AI1 platform.
The duo entered into a collaboration in March 2019 for deploying Zebra Medical Vision’s artificial intelligence (AI) based tools at scale across India with the backing of India-Israel Industrial R&D and Technological Innovation Fund.
The company said that after the deployment of the intracranial hemorrhage, its AI COVID-19 solution will facilitate future collaboration on additional medical findings, including tuberculosis (TB).
Zebra Medical Vision said that its solution, based on a machine learning algorithm which analyzes computed tomography (CT) scans, can help ease the burden on health systems across the world, via earlier diagnoses and more effective triage.
The medical imaging analytics company said that its software provides a quicker alternative for the detection of COVID-19 when PCR testing is not available or delayed, tracking of patients with coronavirus, as decision support for the allocation of ER, ICU resources apart from quantifying the disease burden and its progression.
Dr. Sreenivasa Raju K – CEO and Medical Director at Apollo Radiology International said: “By identifying and tracking the development of these lesions, Zebra-Med software offers key insight into disease severity and enables doctors to diagnose and evaluate patients swiftly and effectively.
“Deciding which suspected COVID-19 patients should be hospitalized is challenging when healthcare systems are overwhelmed. The vast majority of COVID-19 patients are mild, so having this automated tool has helped make critical decisions on whether to hospitalize.”
The artificial intelligence solution for COVID-19 makes use of Zebra Medical Vision’s Ground-Glass Opacities (GGO) algorithm, which is said to be a dominant imaging feature of COVID-19 pneumonia. The Israeli firm said that the algorithm has been modified to include consolidations and offer an automated indication and quantification of suspected COVID-19 findings on standard chest CTs, both contrast and non-contrast, by measuring the percentage of impacted lung-burden volume associated with the entire lung volume, segmentation of the suspected findings, and classification.
Ohad Arazi – CEO of Zebra Medical Vision said: “Zebra-Med researchers and engineers used clinical data from a vast amount of medical images to train the AI model of the COVID-19 solution.
“Our partnership with Apollo Hospitals allows us to leverage a patented algorithm we had developed in the past for imaging features of COVID-19 pneumonia in order to help Apollo’s staff to track the progression of the disease.”